Introduction
Prostate cancer (PCa) is a highly prevalent malignancy worldwide that exhibits a propensity for secondary growth in bone 1 , 2 , 3 . These secondary lesions induce skeletal-related complications, including pain, spinal cord compression, hypercalcemia, diminished mobility, and fractures 4 , 5 , which have a negative impact on patient quality of life and survival 3 .
Radium-223 ( 223 Ra) is an alkaline earth metal radioisotope with properties similar to calcium; upon administration, 223 Ra selectively binds to hydroxyapatite, which is highly enriched in the bone matrix 6 . This incorporation is more pronounced in areas of active bone remodeling, e . g ., at the front of the growth plate or in metastatic lesions 6 . 223 Ra emits alpha particles with high energy but limited tissue penetrance (<100 μm) 7 , which minimizes its impact on healthy bone marrow tissue and reduces the side effects commonly associated with the use of β -emitters 6 , 8 , 9 . In men afflicted with metastatic PCa to bone, 223 Ra prolonged the median overall survival (3.6 months) and delayed the time to first symptomatic skeletal event (5.8 months) compared to placebo 7 , 10 . These results led to the approval of 223 Ra as the first bone-targeted radioligand for the treatment of metastatic PCa patients with symptomatic bone lesions and no visceral involvement 7 , 10 . Since then, 223 Ra has been clinically investigated for other cancer types that colonize the bone, including multiple myeloma, hormone-positive breast cancer, renal cell carcinoma, non-small cell lung cancer, and differentiated thyroid cancer 11 , 12 , 13 .
The efficacy of 223 Ra in patients correlates with a sustained decline in osteoblast-derived alkaline phosphatase (ALP) in serum, denoting reduced bone turnover, which was not associated with increased bone fragility in the ALSYMPCA clinical trial 7 . A recent prospective clinical trial investigating the effects of 223 Ra on bone health by whole-body magnetic resonance imaging (MRI) showed a higher fracture rate, mostly asymptomatic, in patients receiving 223 Ra. Sixty-eight percent of the fractures occurred in bones without metastases 14 , suggesting a systemic rather than tumor-dependent effect on bone biology. However, the mechanisms leading to 223 Ra-associated bone fragility remain unclear, and scarce and contrasting data are available on the effects of 223 Ra on bone stromal cells. In vitro , 223 Ra inhibits osteoblast and osteoclast differentiation (but not bone resorption) at doses >400 Bq/mL and induces their activation at doses <200 Bq/mL 11 . In mouse models of PCa bone metastasis, 223 Ra decreases the number of osteoblasts and osteoclasts in mice implanted with LNCaP, whereas in mice implanted with LuCap58, it only reduces the number of osteoblasts 15 .
External beam radiation is often used in patients with bone tumors, and its effects on bone health are well documented 16 , 17 , 18 , 19 . External beam radiation in rodent models has been shown to induce bone marrow adipogenesis, impaired mechanical properties, altered material composition, increased osteoclastogenesis, trabecular bone loss, and cytotoxic effects on osteoblasts and bone-lining cells 20 , 21 , 22 , 23 . However, 223 Ra fundamentally differs from traditional external beam radiation, i.e. , alpha radiation is constantly emitted from the bone itself with a half-life of 11.4 days and a maximum penetrance of 100 μm 8 . Therefore, the impact of 223 Ra on bone health may differ substantially from that of external beam radiation.
Interestingly, 223 Ra-induced bone fragility can be mitigated through the administration of bone-protecting agents, such as zoledronic acid (ZA), as recently assessed in the EORTC-1333-GUCG trial 24 . Besides evidence of reduced osteoclast numbers via a decrease in TRACP5b circulating levels in tumor-bearing mice treated with 223 Ra and ZA, no further studies have been performed to address the mechanisms underlying such improvements in bone health. A more thorough evaluation of the radiobiological effects of 223 Ra on the bone matrix and cells is thus needed to support a meaningful and safe application of this agent in patients.
In this study, we evaluated the effects of 223 Ra treatment on the bone microenvironment. We combined 3D fluorescence microscopy, micro-computed tomography, mechanical testing, and computational simulations to investigate the impact of 223 Ra on bone, including its effects on bone mechanical properties and resident cell populations, as well as the impact of ZA in mitigating 223 Ra-induced bone damage.
Materials and methods
In vivo studies
Animal studies were approved by the Institutional Animal Care and Use Committee (IACUC) of the University of Texas MD Anderson Cancer Center and were performed according to the institutional guidelines for animal care and handling. Details are provided in the Supporting Information
Statistical analysis
Statistical analysis was performed using GraphPad Prism 10.0. Data are shown as mean ± standard deviation (SD). An unpaired two-sided Student's t test was applied to analyze two populations, whereas one-way ANOVA followed by the Tukey honestly significant difference (HSD) post hoc test, was performed to compare more than two populations. All statistical tests were two-sided, and P values less than 0.05 were considered statically significant. Further experimental methods are detailed in the Supporting Information
Results
Effects of 223 Ra on bone microarchitecture
To monitor the impact of 223 Ra on bone microarchitecture, we performed ex vivo micro-computed tomography (micro-CT) on adult mouse tibiae (8 weeks old, n = 4/group) harvested at 1, 2, 4, 6, and 8 weeks after a single injection of 223 Ra (300 kBq/kg) or vehicle (normal saline) ( Fig. 1 A). Consistent with previous reports on radiation-induced trabecular bone damage 19 , we observed trabecular bone microarchitecture destruction in the proximal tibia metaphysis region as early as 1 week post 223 Ra treatment ( Fig. 1 B and C). Specifically, 223 Ra treatment induced a significant decrease in bone volume fraction, bone surface, and bone mineral density compared to control. Furthermore, 223 Ra induced a significant increase in trabecular separation and decrease in trabecular number ( Fig. 1 B and C). These are both key markers of bone remodeling and reflect an altered state induced by 223 Ra. Tibiae collected 2 weeks post treatment showed further trabecular bone impairment, which stabilized 4 weeks post 223 Ra injection and remained stable up to 8 weeks post injection ( Supporting Information Fig. S1 ). Cortical and trabecular bone thickness did not show significant differences at any timepoint. To confirm these findings in a different skeletal element, micro-CT analysis was performed on lumbar spine samples harvested at 4 weeks. Results showed similar trends, with a significant decrease in key trabecular bone parameters ( Supporting Information Fig. S2 ). Taken together, these data show that 223 Ra impairs trabecular bone architecture but does not affect cortical bone. Figure 1 223 Ra disrupts bone microarchitecture and mineral density. (A) Experimental design. (B) Representative micro-CT micrographs. Bar, 300 μm. (C) Micro-CT analysis of bone structural parameters. Bar, 300 μm. P values by unpaired two-tailed Student's t test, P < 0.05 (∗); P < 0.01 (∗∗); P < 0.0001 (∗∗∗). Figure 1
Effects of 223 Ra on bone mechanical properties
To assess the consequences of 223 Ra on bone mechanical properties, we performed three-point bending test analysis on mouse femurs ( n = 5 to 8/group). Femurs were harvested at 2, 4, and 8 weeks post 223 Ra (300 kBq/kg) or vehicle (saline) injection ( Fig. 2 A and B), and mechanical testing was performed. The stiffness and rigidity of the femurs were slightly, but significantly, increased for the 223 Ra group at 2 weeks as compared to control ( Fig. 2 C). The ultimate load, ultimate strength, elastic modulus, and cross-sectional area parameters did not show any significant differences ( Fig. 2 C). Taken together, these data suggest that 223 Ra treatment has very limited and transient effects on the mechanical properties of murine long bones. However, the three-point bending test mainly provides information on cortical bone mechanical properties (not trabecular bone), which were not expected to significantly change based on micro-CT analysis (see Fig. 1 B and C). To investigate the mechanical properties of trabecular bone, we used tibial micro-CT trabecular bone data (2, 4 and 8 weeks) to develop a microstructural finite element analysis (m-FEA) of a central cubic volume (0.5 mm × 0.5 mm × 0.5 mm) of trabecular bone ( Fig. 2 D and E). m-FEA is a well-established method to study the mechanical behavior of trabecular bone under loading conditions 25 , 26 , 27 . By simulating a uniaxial compression test, m-FEA results predicted a significant, progressive decrease in stiffness and increased deformation for the 223 Ra-treated group compared to the control group at 4 and 8 weeks post treatment ( Fig. 2 D). A significant increase in both maximum Von Mises stress (a measure of the overall stress state of the material under compressive loading) and strain (a measure of the deformation of the material with respect to the applied load) was also observed in the 223 Ra-treated group compared to the control group at both 4 and 8 weeks ( Fig. 2 D). These results were further coupled with a significant rise in the maximum strain energy and in the ratio of total energy per number of trabeculae at both 4 and 8 weeks, in alignment with a reduced number of trabeculae ( Fig. 2 D). In addition, no significant differences were observed in the effective elastic modulus, with values closely matching the initial micro-CT-derived elastic modulus ( E = 1695 MPa). This is in line with the expectation that, within the linear elastic regime, stress and strain remain proportional and, consequently, their ratio also remains unchanged. Altogether, these results suggest that 223 Ra impacts the mechanical properties of trabecular bone but not of cortical bone. Figure 2 Effects of 223 Ra on bone mechanical properties. (A) Experimental design. (B, C) 3-point bending test setup and mechanical analysis. (D) m-FEA analysis; representative images (micro-CT-derived volume, deformation and Von Mises stress maps at 4 weeks) with quantifications are shown. Bar, 500 μm. P values by unpaired two-tailed Student's t test, P < 0.05 (∗); P < 0.01 (∗∗); P < 0.001 (∗∗∗). Figure 2
To evaluate the impact of 223 Ra on trabecular bone composition, we performed confocal Raman microscopy on the condyle region of mouse femurs ( n = 7 to 17 scans/group). For this purpose, bones were harvested at two timepoints (2 and 4 weeks) after a single injection of 223 Ra (300 kBq/kg) or vehicle (saline). Raman spectra analysis showed a significant increase in collagen crosslinking 4 weeks after 223 Ra treatment as compared to control. Interestingly, this increase was coupled with a significant decrease in collagen density at the same timepoint ( Supporting Information Fig. S3 ). Taken together, confocal Raman analysis on femoral trabecular bone suggests a direct impact of 223 Ra on the degree of collagen crosslinking and collagen density in the bone matrix.
Direct impact of radiation vs. cell mediated effects of 223 Ra on bone microarchitecture
To determine whether the deleterious effects of 223 Ra on bone microarchitecture are predominantly caused by the direct impact of radiation on mineralized bone tissue or are mediated by the cells, mouse tibiae were harvested 2 days after a single injection of 223 Ra (300 kBq/kg) or vehicle (saline) ( Supporting Information Fig. S4A ). To account for different preservation methods that could impact the effects of 223 Ra therapy, tibiae were either fixed with 70% ethanol and kept at 4 °C, or frozen and kept at −20 °C without fixation. Ex vivo micro-CT analysis was then performed at 1, 2, and 4 weeks after 223 Ra or vehicle injection. Contrary to our previous findings, bone microarchitecture was stable and did not show any significant changes in bone parameters across different timepoints regardless of the sample preservation method ( Fig. S4B and S4C ). Taken together, these micro-CT data suggest that the trabecular bone loss induced by 223 Ra does not simply rely on direct damage inflicted by bone-targeted alpha radiation but is largely cell mediated.
Effects of 223 Ra on osteoblasts in vivo and in vitro
To monitor the in vivo effects of 223 Ra on osteoblasts (bone-forming cells) , we used a fluorescent reporter mouse model expressing mCherry fluorochrome under the control of the osterix (Osx) promoter. Osx, also called Sp7, is a cell-specific transcription factor expressed by mature osteoblasts and osteocytes that regulates their differentiation 28 . Sp7-mCherry mice were treated with 223 Ra (300 kBq/kg) or vehicle and euthanized at different timepoints up to 60 days. Tibiae were harvested, decalcified, and sectioned into thick slices (300 μm) using a vibratome. Slices from the tibial metaphysis region were stained with DAPI (nuclear counterstain) and an anti-alkaline phosphatase (ALP) antibody ( Fig. 3 A). ALP, an early osteogenic marker of bone formation, is a membrane-associated glycoprotein that increases local concentrations of inorganic phosphate and facilitates bone mineralization 29 . Images were acquired via confocal microscopy, and image analysis was performed using FIJI 30 to quantify the impact of 223 Ra on bone coverage of Osx + cells and ALP + cells ( Fig. 3 B). Immunofluorescence analysis showed a significant decrease in the percentage of Osx + cells lining the bone surface on Days 7 (−13 ± 5%) and 11 (−25 ± 7%) post 223 Ra treatment, which was overcome on Day 18 ( Fig. 3 B). Interestingly, the amount of perivascular Osx + precursors within the bone marrow significantly increased after 7 days and then decreased over time ( Fig. 3 C). Figure 3 Effects of 223 Ra on osteoblasts in vivo . (A) Experimental design; (i) immunofluorescence; (ii) dual bone labelling. (B) Osx and ALP, representative images and quantification ( n = 5) in tibiae at different timepoints post 223 Ra treatment. Bar, 100 μm. (C) Osx + cells into the bone marrow, representative images and quantifications. Bar, 50 μm. (D, E) ALP + layer thickness, representative images and quantification (altered co-localization of osx + cells is shown). Bar, 30 μm. (F) Representative images and quantification of mineral apposition of femoral sections (green, calcein green; red, xylenol orange; n = 6). Bar, 50 μm. P values by one-way ANOVA with Tukey's HSD post hoc test; P < 0.05 (∗); P < 0.01 (∗∗); P < 0.001 (∗∗∗). Figure 3
No significant change in ALP + cell coverage was detected up to Day 60. However, analysis of the bone interface showed a significant thickening of the ALP + layer on Days 7 and 11 post 223 Ra treatment (with a maximum extension of 20.07 ± 7.23 μm from the bone interface towards the bone marrow cavity on Day 7), and significantly decreased by Day 18 ( Fig. 3 D). Furthermore, as compared to control, the ALP + lining showed an irregular morphology and no colocalization with Osx + signal ( Fig. 3 D and E).
To assess the degree of osteoblast function and bone mineralization in vivo after 223 Ra treatment, we measured a dynamic index of bone formation following subsequent injections of two calcium-labeling agents, calcein green and xylenol orange ( Fig. 3 A and F). We observed no difference in calcium labeling on the femur cortical bone region, consistent with the absence of structural phenotype in this compartment. In the trabecular bone region of the femoral secondary spongiosa, however, the mineral apposition rate (MAR) was significantly decreased in 223 Ra-treated mice as compared to the control group ( Fig. 3 F). Similarly, in vitro osteoblast cultures showed significantly decreased calcium deposition upon 223 Ra treatment, as monitored by calcein blue staining ( Supporting Information Fig. S5 ). Overall, these results suggest that 223 Ra affects osteoblast activity.
Effect of 223 Ra on osteoclasts in vivo
To monitor the in vivo effects of 223 Ra on osteoclasts (bone-resorbing cells), immunofluorescence experiments on bone slices were performed using an anti-tartrate-resistant acid phosphatase (TRAP) antibody. TRAP is an enzyme secreted by osteoclasts and a key effector of their bone resorption activity 31 . Samples were acquired via confocal microscopy, and image analysis was performed to quantify the bone coverage of TRAP + cells at different timepoints ( Fig. 4 A). Image analysis showed a significant, transient increase in TRAP + cell bone coverage 2 days after 223 Ra treatment, followed by a significant reduction in TRAP + cells between Days 18 and 40 post treatment, and a recovery on Day 60 ( Fig. 4 B). These results are in line with results obtained after external beam radiation 20 . Serum analysis of circulating tartrate-resistant acid phosphatase 5b (TRACP5b) confirmed the transient increase in osteoclast number identified by immunofluorescence 2 days post 223 Ra treatment and the subsequent drop below baseline ( Fig. 4 C). Overall, our immunofluorescence analysis coupled with bone turnover serum marker analysis showed a transient osteoclastogenic effect induced by 223 Ra, followed by a drop in osteoclast number and activity. Figure 4 Effect of 223 Ra on osteoclasts in vivo . (A, B) TRAP + cells, representative images and quantification at different timepoints post 223 Ra treatment. Inserts, magnification of TRAP + cell-bone interface. Bar, 100 μm. (C) Serum levels of TRACP5b by ELISA. P values by one-way ANOVA with Tukey's HSD post hoc test; P < 0.01 (∗∗); P < 0.001 (∗∗∗). Figure 4
Effect of 223 Ra on blood vessels in vivo and in vitro
Next, we investigated the impact of 223 Ra on blood vessel biology in vivo. Besides delivering oxygen and nutrients within the bone marrow compartment, bone blood vessels have been implicated in supporting the function of both the hematopoietic and bone stromal compartments 32 . The density of blood vessels within the bone marrow cavity was assessed by immunofluorescence analysis using an anti-endomucin antibody, a protein expressed in bone sinusoidal blood vessels. Quantitative analysis showed that 223 Ra induced a significant, transient increase in the overall amount of bone marrow blood vessels 7 days after treatment, which returned to baseline on Day 11 post treatment ( Fig. 5 A). Since 223 Ra exerts space-dependent cytotoxic effects due to the short penetrance of alpha particles (<100 μm), we further performed a differential analysis of its impact on blood vessels. We selected 4 subregions of the bone marrow, based on their distance from bone, at 0 to 20 (area that showed maximum ALP + positivity upon exposure to 223 Ra, see Figs. 3 D), 20 to 100, 100 to 200 and > 200 μm. The amount of blood vessels proximal to the bone interface (<20 μm) was significantly reduced 7 days after 223 Ra dosing ( Fig. 5 B), consistent with the transient occupation of this area by ALP + cells ( Fig. 5 C) but returned to basal levels by Day 11. Vessels at 20 to 100, 100 to 200 and > 200 μm distance, instead, all showed a significant increase on Day 7 ( Fig. 5 A). In addition, we generated a 3D HUVECs vascular network within a hydrogel system and treated it with 800 Bq/mL 223 Ra and monitored the therapy response via live immunofluorescence for up to 2 weeks. No significant difference was identified in GFP signal across the different timepoints between control and treated groups, even initially when radiation was higher, confirming that 3D blood vessel structures are resistant to radiation ( Fig. 5 D). Additionally, quantification of adipocyte infiltration was performed by counting their number based on their characteristic morphology (a spherical shape and a diameter of >30 μm, distinguishable from endomucin + blood vessels, characterized by an elongated, irregular shape and frequent interconnections). This morphology-based identification was further validated by adipocyte-specific marker staining and histology ( Supporting Information Fig. S6 ). Results showed a significant increase in adipocyte infiltration 18 days after 223 Ra dosing, with a sustained increase up to day 60 ( Fig. 5 E). Bone marrow adipocyte infiltration is a known effect of 223 Ra treatment, as previously shown 33 . Altogether, these data confirm the induction of adipocytes and suggest that 223 Ra does not exert major toxic effects on bone blood vessels. Figure 5 Effects of 223 Ra on blood vessels in vivo and in vitro . (A) Endomucin (emcn), representative images and quantification in tibiae at different timepoints post 223 Ra treatment. Box, magnifications shown in (E). Bar, 100 μm. (B) Differential analysis of blood vessel coverage. Bar, 20 μm. (C) Impact of ALP thickening on blood vessel coverage. (D) GFP + HUVEC 3D vascular network treated with different doses of 223 Ra, representative images and quantification. Bar, 50 μm. (E) Bone marrow adipocyte infiltration, representative images and quantification. Dashed blue line with asterisk, adipocyte. Bar, 100 μm. P values by one-way ANOVA with Tukey's HSD post hoc test; P < 0.05 (∗); P < 0.01 (∗∗); P < 0.001 (∗∗∗). Figure 5
Effects of 223 Ra+ZA combination treatment on bone properties
ZA is a drug widely used as bone-protecting agent in patients with bone metastatic cancer and other bone degenerative diseases 34 . ZA has high affinity for hydroxyapatite 35 , inhibits osteoclast resorptive activity, and possibly targets other bone stromal cells, including adipocytes and osteoblasts 36 , 37 , 38 . Current clinical guidelines recommend giving ZA to patients receiving 223 Ra therapy, as it has been shown to decrease 223 Ra-induced bone fragility 39 . However, the biological determinants of this response have never been addressed.
To evaluate the effects of combination 223 Ra + ZA treatment on bone microarchitecture, we performed ex vivo micro-CT analysis of mouse tibiae harvested 4 weeks post treatment. This analysis showed bone-protective effects of ZA on 223 Ra-mouse bones and confirmed our previous findings on trabecular bone damage by 223 Ra monotherapy. Specifically, combination 223 Ra + ZA prevented a decrease in bone volume fraction and trabecular number ( Fig. 6 C). Combination treatment also prevented an increase in trabecular separation as compared to 223 Ra monotherapy. ZA monotherapy and combination therapy also resulted in a significant increase in trabecular bone thickness. As observed with 223 Ra monotherapy, ZA did not induce significant changes in cortical bone thickness ( Fig. 6 A‒C). Taken together, micro-CT data confirmed the deleterious effects of single-agent 223 Ra and highlighted the importance of using bone-protecting agents like ZA to prevent radiation-induced trabecular bone damage. Figure 6 ZA blocks 223 Ra-induced deleterious effects on trabecular bone microarchitecture. (A‒C) Representative micro-CT images and quantification at 4-week time point. Bar, 300 μm, n = 4 tibiae/group. P values by one-way ANOVA with Tukey's HSD post hoc test; P < 0.01 (∗∗); P < 0.001 (∗∗∗). Figure 6
Effects of 223 Ra + ZA combination treatment in vivo
To investigate the impact of ZA + 223 Ra combination treatment on the bone cellular components, immunofluorescence experiments on bone slices were performed as described above in 223 Ra monotherapy experiments. Mice were treated with 1 dose of 223 Ra and with ZA weekly, for 4 weeks ( Fig. 7 A). ZA induced a significant reduction in osteoclast coverage after 4 weeks, with or without 223 Ra ( Fig. 7 B). This is in line with one of the proposed modes of action of ZA, which has been shown to reduce osteoclast number and activity. To determine whether ZA had any impact on the osteoblast population, tibiae from sp7-mcherry mice were harvested at 7 and 11 days post treatment, based on the timing of the Osx + cell reduction observed with 223 Ra monotherapy ( Fig. 3 B). ZA had no effect on Osx + lining cells and did not affect the Osx + cell reduction 7 days post 223 Ra treatment (−9 ± 3%; Fig. 7 C); however, ZA completely overcame the major drop in Osx + cells observed 11 days after 223 Ra administration (−25 ± 7%; Fig. 3 B). In addition, ZA or combination therapy did not affect ALP bone coverage across the different timepoints. However, when 223 Ra was administered in the presence of ZA, the thickening of the ALP + cell layer was not observed ( Fig. 7 D). Additionally, adipocyte number was quantified 4 weeks post treatment. While ZA treatment alone did not impact adipocyte number ( Fig. 7 E), when ZA was combined with 223 Ra, significantly fewer adipocytes were observed in the bone marrow as compared to 223 Ra monotherapy. Figure 7 ZA minimizes osteoblast damage and decreases osteoclast number in vivo . (A) Experimental design. (B) TRAP + cells, representative images and quantification of tibiae harvested 4 weeks after ZA or ZA + 223 Ra treatment. Inserts, magnification of TRAP + cell–bone interface. Bar, 100 μm. (C) Osx and ALP, representative images and quantification in tibiae. Bar, 100 μm. (D) ALP + layer thickness, representative images and quantification. Bar, 30 μm. (E) High magnification inserts (from panel C) and quantification of bone marrow adipocyte infiltration. Dashed blue line with asterisk, adipocyte; Bar, 50 μm. P values by one-way ANOVA with Tukey's HSD post hoc test; P < 0.05 (∗); P < 0.01 (∗∗); P < 0.001 (∗∗∗). Figure 7
Taken together, our findings highlight the importance of using bone-protecting agents to prevent radiation-induced trabecular bone loss and suggest that, besides canonical osteoclast targeting, ZA might improve 223 Ra-induced bone fragility by directly or indirectly affecting MSC-derived cell (osteoblasts, adipocytes) biology.
Discussion and conclusions
In this study, we investigated the effects of 223 Ra on the skeleton and key bone stromal components, including osteoblasts, osteoclasts, blood vessels, and adipocytes. 223 Ra therapy had a deleterious effect on the trabecular bone compartment of both long bones (tibiae) and lumbar vertebrae, but not on cortical bone. This may be due to differences in bone composition and remodeling rates between the two compartments, which in turn can affect local incorporation of 223 Ra. 223 Ra-induced trabecular bone loss was rapid, evident as early as 1 week post treatment, and resembled age-related osteoporosis 40 . Three-point bending tests did not show major effects of 223 Ra on cortical bone mechanical properties, as expected based on our micro-CT data, while m-FEA and simulations of uniaxial compression predicted a significant impact on trabecular mechanical properties. Specifically, stiffness decreased, while deformation, Von Mises stress and strain, and strain energy increased. These computational analyses allowed us to specifically test the mechanical properties of the trabecular region, which would have otherwise been difficult to test ex vivo due to the small size of the mouse bones and limited ability to discern the contributions of the cortical versus trabecular compartments to the overall bone mechanical properties.
Collectively, our results showed that 223 Ra exerts a major impact on trabecular bone and suggested that vertebrae should be more susceptible to 223 Ra-induced bone fragility 41 due to a higher trabecular content as compared to long bones. In further support of these findings, fractures in patients were primarily detected in the spine (66%), followed by the thorax (25%) and pelvis (6%) 14 , when whole-body MRI was specifically applied after 223 Ra treatment. Long bones, on the other hand, were minimally affected by 223 Ra (∼1% of fractures) 14 . Overall, it is possible that the majority of 223 Ra-induced fractures are in the spine and are clinically insignificant. This would explain why the ALSYMPCA clinical trial showed no significant increase in bone fractures following treatment with 223 Ra (4% in patients receiving 223 Ra compared to 5% in those receiving a placebo).
External beam radiation induces trabecular bone loss and other changes in the bone microenvironment through cellular and non-cellular mechanisms 16 , 20 , 22 . To determine whether 223 Ra inflicted direct, non-cell-mediated bone damage, we performed ex vivo micro-CT analysis on bone harvested 2 days post treatment. In the absence of vital cells, 223 Ra treatment did not affect bone microarchitecture or trabecular bone volume, suggesting that 223 Ra causes changes in the bone microenvironment predominantly by cell-mediated mechanisms rather than through direct damage to the mineralized tissue.
Bone remodeling and homeostasis are closely regulated by the coupled action of osteoblasts and osteoclasts. We first focused on understanding the impact of 223 Ra on osteoblasts, the bone-forming component. Immunofluorescence analysis showed a significant decrease in the percentage of Osx + cells lining the bone surface on Days 7 and 11 post 223 Ra treatment; however, no significant reduction in ALP + cell coverage was identified up to Day 60. Interestingly, local ALP expression expanded across 2 to 3 layers of Osx ‒ cells in 223 Ra-treated mice, likely indicating a perturbation of osteoblast maturation within a few days of 223 Ra administration. We identified an increase in perivascular Osx + precursors within the bone marrow 7 days post 223 Ra treatment, which decreased to basal levels within 30 days, likely representing osteoprogenitors stimulated by 223 Ra-induced damage and involved in repair. As compared to mature osteoblasts, this proliferating cell population might be more efficiently targeted by 223 Ra, contributing to the impairment observed in our experiments. We observed no difference in bone formation in cortical bone regions, whereas mineral apposition rate in bone trabeculae was significantly decreased in 223 Ra-treated mice as compared to the control group, indicating that 223 Ra affected osteoblast bone-forming activity. In addition to partially inducing bone fragility, impaired osteoblast function might explain the reduced osteoblastic activity (circulating ALP levels) observed in patients with PCa and osteoblastic bone metastasis treated with 223 Ra 7 .
The results of our investigation into the effects of 223 Ra on osteoclasts (the bone-resorbing component) showed a transient increase in TRAP + cell bone coverage 2 days after 223 Rad treatment, followed by a significant decrease in TRAP + cells by Day 18 post 223 Ra treatment, and a recovery of these cells by day 60, in line with published studies involving external beam radiation 20 . The transient spike in osteoclast number was confirmed by serum analysis of TRACP5b, an osteoclast-specific marker. Taken together, these results suggest the trabecular bone loss induced by 223 Ra is mediated by a rapid osteoclastic response coupled to dysfunctional bone deposition by osteoblasts.
We also observed a significant increase in adipocyte number as early as 18 days post 223 Ra treatment. Adipogenesis and adipocyte infiltration are well-characterized side effects of external radiation, and our data suggest that alpha-particle radiation has a similar effect 42 . Interestingly, bone marrow adipocytes have been shown to negatively control bone mass mediated by bone morphogenetic protein receptor activation 43 . Therefore, increased adipocyte formation might play a role in impaired osteoblast function. As an alternative mechanism, commitment of MSCs toward an adipocyte lineage may further reduce osteoblastogenesis 44 .
223 Ra induced a significant increase in the overall quantity of bone marrow blood vessels 7 days after treatment, which returned to baseline levels 11 days post treatment. We previously demonstrated that 223 Ra has a transient cytotoxic effect on hematopoietic cells, which peaks on Day 4 post treatment, followed by a recovery phase evident on Day 11 33 . Consistently, the loss of hematopoietic cells in response to external beam radiation triggers vessel dilation, permeability, and endothelial cell proliferation 45 . Additionally, our differential analysis showed that the amount of blood vessels proximal to the bone interface (<20 μm; the same area affected by ALP thickening) was significantly reduced at 7 days after 223 Ra dosing but returned to baseline levels on Day 11. In vitro experiments with 3D vascular networks did not show any 223 Ra-mediated reduction of these vascular structures, suggesting that 223 Ra does not exert direct toxicity on blood vessels.
ZA is commonly used as a bone-protecting agent in the clinical setting, and its administration to 223 Ra patients showed positive effects on bone fragility 24 . Our results confirmed the bone-protective properties of ZA on the trabecular bone damage induced by 223 Ra, likely via its inhibitory effect on bone resorption. Interestingly, ZA significantly decreased 223 Ra-induced bone marrow adipocyte density, loss of Osx + cells, and impact on ALP expression; it remains to be determined whether this occurs via direct action on MSCs, osteoblasts, or adipocytes or as an indirect effect secondary to reduced bone turnover. Nevertheless, these results confirm the importance of using ZA in combination with 223 Ra therapy to prevent trabecular bone loss and reduce the overall impact of 223 Ra on the bone cell population, as recommended by all major medical guidelines. Comparison with other anti-resorptive agents with convergent and divergent mechanisms of action ( i.e. , denosumab, which inhibits osteoclast maturation) is needed to identify the best option to rationally prevent 223 Ra-induced bone fragility.
Author contributions
Sergio Barrios, Elisa Serafini, Nicholas J. Dunbar, Paul Corn, Florent Elefteriou, Catherine Ambrose, Stefano Casarin, Antonio G. Mikos, Eleonora Dondossola designed the research. Sergio Barrios, Elisa Serafini, Ludovica La Posta, D. Nicole Meyers, Eleonora Dondossola carried out the experiments and performed data analysis. Nicholas J. Dunbar, Florent Elefteriou, Catherine G. Ambrose, Stefano Casarin participated part of the experiments. Sergio Barrios, Elisa Serafini, Stefano Casarin, Antonios G. Mikos, and Eleonora Dondossola wrote the manuscript. Ludovica La Posta, D. Nicole Meyers, Nicholas J. Dunbar, Paul G. Corn, Florent Elefteriou, Catherine G. Ambrose revised the manuscript. All the authors have read and approved the final manuscript.
Conflicts of interest
No potential conflicts of interest relevant to this article exist. The funders had no role in the design of the study, the collection/analysis/interpretation of data, or the decision to submit the manuscript.
References
1 Torre L.A. Bray F. Siegel R.L. Ferlay J. Lortet-Tieulent J. Jemal A. Global cancer statistics, 2012 CA Cancer J Clin 65 2015 87 108 25651787 10.3322/caac.21262
2 Weilbaecher K.N. Guise T.A. McCauley L.K. Cancer to bone: a fatal attraction Nat Rev Cancer 11 2011 411 425 21593787 10.1038/nrc3055 PMC3666847
3 Croucher P.I. McDonald M.M. Martin T.J. Bone metastasis: the importance of the neighbourhood Nat Rev Cancer 16 2016 373 386 27220481 10.1038/nrc.2016.44
4 Woodward E. Jagdev S. McParland L. Clark K. Gregory W. Newsham A. Skeletal complications and survival in renal cancer patients with bone metastases Bone 48 2011 160 166 20854942 10.1016/j.bone.2010.09.008
5 Yuasa T. Urakami S. Yamamoto S. Yonese J. Saito K. Takahashi S. Treatment outcome and prognostic factors in renal cell cancer patients with bone metastasis Clin Exp Metastasis 28 2011 405 411 21365325 10.1007/s10585-011-9379-7
6 Abou D.S. Ulmert D. Doucet M. Hobbs R.F. Riddle R.C. Thorek D.L. Whole-body and microenvironmental localization of radium-223 in naive and mouse models of prostate cancer metastasis J Natl Cancer Inst 108 2016 [页码] 10.1093/jnci/djv380 PMC4849807 26683407
7 Parker C. Nilsson S. Heinrich D. Helle S.I. O'Sullivan J.M. Fossa S.D. Alpha emitter radium-223 and survival in metastatic prostate cancer N Engl J Med 369 2013 213 223 23863050 10.1056/NEJMoa1213755
8 Bruland O.S. Nilsson S. Fisher D.R. Larsen R.H. High-linear energy transfer irradiation targeted to skeletal metastases by the alpha-emitter 223Ra: adjuvant or alternative to conventional modalities?. Clin Cancer Res 12 2006 6250s-7s 10.1158/1078-0432.CCR-06-0841 17062709
9 Dondossola E. Casarin S. Paindelli C. De-Juan-Pardo E.M. Hutmacher D.W. Logothetis C.J. Radium 223-mediated zonal cytotoxicity of prostate cancer in bone J Natl Cancer Inst 111 2019 1042 1050 30657953 10.1093/jnci/djz007 PMC6792112
10 Sartor O. Coleman R. Nilsson S. Heinrich D. Helle S.I. O'Sullivan J.M. Effect of radium-223 dichloride on symptomatic skeletal events in patients with castration-resistant prostate cancer and bone metastases: results from a phase 3, double-blind, randomised trial Lancet Oncol 15 2014 738 746 24836273 10.1016/S1470-2045(14)70183-4
11 Suominen M.I. Rissanen J.P. Kakonen R. Fagerlund K.M. Alhoniemi E. Mumberg D. Survival benefit with radium-223 dichloride in a mouse model of breast cancer bone metastasis J Natl Cancer Inst 105 2013 908 916 23682134 10.1093/jnci/djt116
12 Ueno N.T. Tahara R.K. Fujii T. Reuben J.M. Gao H. Saigal B. Phase II study of Radium-223 dichloride combined with hormonal therapy for hormone receptor-positive, bone-dominant metastatic breast cancer Cancer Med 9 2020 1025 1032 31849202 10.1002/cam4.2780 PMC6997080
13 Morris M.J. Corey E. Guise T.A. Gulley J.L. Kevin Kelly W. Quinn D.I. Radium-223 mechanism of action: implications for use in treatment combinations Nat Rev Urol 16 2019 745 756 31712765 10.1038/s41585-019-0251-x PMC7515774
14 Hijab A. Curcean S. Tunariu N. Tovey H. Alonzi R. Staffurth J. Fracture risk in men with metastatic prostate cancer treated with radium-223 Clin Genitourin Cancer 19 2021 e299 e305 33958296 10.1016/j.clgc.2021.03.020 PMC8514085
15 Suominen M.I. Fagerlund K.M. Rissanen J.P. Konkol Y.M. Morko J.P. Peng Z. Radium-223 inhibits osseous prostate cancer growth by dual targeting of cancer cells and bone microenvironment in mouse models Clin Cancer Res 23 2017 4335 4346 28364014 10.1158/1078-0432.CCR-16-2955 PMC5540794
16 Green D.E. Rubin C.T. Consequences of irradiation on bone and marrow phenotypes, and its relation to disruption of hematopoietic precursors Bone 63 2014 87 94 24607941 10.1016/j.bone.2014.02.018 PMC4005928
17 Oest M.E. Policastro C.G. Mann K.A. Zimmerman N.D. Damron T.A. Longitudinal effects of single hindlimb radiation therapy on bone strength and morphology at local and contralateral sites J Bone Miner Res 33 2018 99 112 28902435 10.1002/jbmr.3289 PMC5776033
18 Costa S. Reagan M.R. Therapeutic irradiation: consequences for bone and bone marrow adipose tissue Front Endocrinol 10 2019 587 10.3389/fendo.2019.00587 PMC6727661 31555210
19 Zhang J. Qiu X. Xi K. Hu W. Pei H. Nie J. Therapeutic ionizing radiation induced bone loss: a review of in vivo and in vitro findings Connect Tissue Res 59 2018 509 522 29448860 10.1080/03008207.2018.1439482
20 Green D.E. Adler B.J. Chan M.E. Lennon J.J. Acerbo A.S. Miller L.M. Altered composition of bone as triggered by irradiation facilitates the rapid erosion of the matrix by both cellular and physicochemical processes PLoS One 8 2013 e64952 10.1371/journal.pone.0064952 PMC3669258 23741433
21 Bartlow C.M. Mann K.A. Damron T.A. Oest M.E. Altered mechanical behavior of demineralized bone following therapeutic radiation J Orthop Res 39 2021 750 760 32965711 10.1002/jor.24868 PMC8212945
22 Green D.E. Adler B.J. Chan M.E. Rubin C.T. Devastation of adult stem cell pools by irradiation precedes collapse of trabecular bone quality and quantity J Bone Miner Res 27 2012 749 759 22190044 10.1002/jbmr.1505
23 Chandra A. Wang L. Young T. Zhong L. Tseng W.J. Levine M.A. Proteasome inhibitor bortezomib is a novel therapeutic agent for focal radiation-induced osteoporosis FASEB J 32 2018 52 62 28860152 10.1096/fj.201700375R PMC5731129
24 Gillessen S. Choudhury A. Rodriguez-Vida A. Nole F. Diaz E.G. Roumeguere T.A. Decreased fracture rate by mandating bone protecting agents in the EORTC 1333/PEACEIII trial combining Ra223 with enzalutamide versus enzalutamide alone: an updated safety analysis J Clin Oncol 39 2021 5002
25 Parr W. Chamoli U. Jones A. Walsh W. Wroe S. Finite element micro-modelling of a human ankle bone reveals the importance of the trabecular network to mechanical performance: new methods for the generation and comparison of 3D models J Biomech 46 2013 200 205 23218138 10.1016/j.jbiomech.2012.11.011
26 Bevill G. Keaveny T.M. Trabecular bone strength predictions using finite element analysis of micro-scale images at limited spatial resolution Bone 44 2009 579 584 19135184 10.1016/j.bone.2008.11.020
27 Oftadeh R. Perez-Viloria M. Villa-Camacho J.C. Vaziri A. Nazarian A. Biomechanics and mechanobiology of trabecular bone: a review J Biomech Eng 137 2015 108021 1080215 10.1115/1.4029176 PMC5101038 25412137
28 Sinha K.M. Zhou X. Genetic and molecular control of osterix in skeletal formation J Cell Biochem 114 2013 975 984 23225263 10.1002/jcb.24439 PMC3725781
29 Yeniyol S. Ricci J.L. Alkaline phosphatase levels of murine pre-osteoblastic cells on anodized and annealed titanium surfaces Eur Oral Res 52 2018 12 19 30574594 10.26650/eor.2018.78387 PMC6300123
30 Schindelin J. Arganda-Carreras I. Frise E. Kaynig V. Longair M. Pietzsch T. Fiji: an open-source platform for biological-image analysis Nat Methods 9 2012 676 682 22743772 10.1038/nmeth.2019 PMC3855844
31 Blumer M.J. Hausott B. Schwarzer C. Hayman A.R. Stempel J. Fritsch H. Role of tartrate-resistant acid phosphatase (TRAP) in long bone development Mech Dev 129 2012 162 176 22579636 10.1016/j.mod.2012.04.003 PMC3419267
32 Hendriks M. Ramasamy S.K. Blood vessels and vascular niches in bone development and physiological remodeling Front Cell Dev Biol 8 2020 [页码] 10.3389/fcell.2020.602278 PMC7729063 33330496
33 Parlani M. Boccalatte F. Yeaton A. Wang F. Zhang J. Aifantis I. (223)Ra induces transient functional bone marrow toxicity J Nucl Med 63 2022 1544 1550 35177425 10.2967/jnumed.121.263310 PMC9536707
34 Polascik T.J. Mouraviev V. Zoledronic acid in the management of metastatic bone disease Therapeut Clin Risk Manag 4 2008 261 268 10.2147/tcrm.s2707 PMC2503661 18728715
35 Saad F. Gleason D.M. Murray R. Tchekmedyian S. Venner P. Lacombe L. Long-term efficacy of zoledronic acid for the prevention of skeletal complications in patients with metastatic hormone-refractory prostate cancer J Natl Cancer Inst 96 2004 879 882 15173273 10.1093/jnci/djh141
36 Jin J. Wang L. Wang X-k Lai P-l Huang M-j Jin D-d Risedronate inhibits bone marrow mesenchymal stem cell adipogenesis and switches RANKL/OPG ratio to impair osteoclast differentiation J Surg Res 180 2013 e21 e29 22487394 10.1016/j.jss.2012.03.018
37 Oryan A. Sahvieh S. Effects of bisphosphonates on osteoporosis: focus on zoledronate Life Sci 264 2021 118681 10.1016/j.lfs.2020.118681 33129881
38 Maruotti N. Corrado A. Neve A. Cantatore F.P. Bisphosphonates: effects on osteoblast Eur J Clin Pharmacol 68 2012 1013 1018 22318756 10.1007/s00228-012-1216-7
39 Cheer S.M. Noble S. Zoledronic acid Drugs 61 2001 799 805 11398911 10.2165/00003495-200161060-00010
40 Shim J. Iwaya C. Ambrose C.G. Suzuki A. Iwata J. Micro-computed tomography assessment of bone structure in aging mice Sci Rep 12 2022 8117 35581227 10.1038/s41598-022-11965-4 PMC9114112
41 Osterhoff G. Morgan E.F. Shefelbine S.J. Karim L. McNamara L.M. Augat P. Bone mechanical properties and changes with osteoporosis Injury 47 Suppl 2 2016 S11 S20 10.1016/S0020-1383(16)47003-8 PMC4955555 27338221
42 Zou Q. Hong W. Zhou Y. Ding Q. Wang J. Jin W. Bone marrow stem cell dysfunction in radiation-induced abscopal bone loss J Orthop Surg Res 11 2016 3 26739584 10.1186/s13018-015-0339-9 PMC4704383
43 Zou W. Rohatgi N. Brestoff J.R. Li Y. Barve R.A. Tycksen E. Ablation of fat cells in adult mice induces massive bone gain Cell Metab 32 2020 801-13.e6 10.1016/j.cmet.2020.09.011 PMC7642038 33027637
44 Muruganandan S. Govindarajan R. Sinal C.J. Bone marrow adipose tissue and skeletal health Curr Osteoporos Rep 16 2018 434 442 29855795 10.1007/s11914-018-0451-y PMC6239196
45 Chen Q. Liu Y. Jeong H.-W. Stehling M. Dinh V.V. Zhou B. Apelin+ endothelial niche cells control hematopoiesis and mediate vascular regeneration after myeloablative injury Cell Stem Cell 25 2019 768-83.e6 10.1016/j.stem.2019.10.006 PMC6900750 31761723
Supplementary data
The following is the Supporting Information to this article: Multimedia component 1 Multimedia component 1
Acknowledgments
This work was supported by 10.13039/501100000801 Bayer HealthCare Pharmaceuticals Inc. (57440); 10.13039/100000005 Department of Defense , CDMRP-KCRP (KC210132P1, USA ); University of Texas MD Anderson Cancer Center; 10.13039/100004917 Cancer Prevention and Research Institute of Texas (RP230160, USA ); 10.13039/100000002 National Institutes of Health (P41EB023833; P30CA016672, USA ). We thank Dr. Ralph Adams (Max Planck Institute for Molecular Biomedicine, Münster, Germany) for sharing mCherry-Sp7 (Osx + ) mice. We thank Sarah Townsend (University of Texas MD Anderson Cancer Center, TX, USA) for careful editing of our paper.
This article is part of a special issue entitled: Targeted Radiopharmaceuticals published in Acta Pharmaceutica Sinica B.
Peer review under the responsibility of Chinese Pharmaceutical Association and Institute of Materia Medica, Chinese Academy of Medical Sciences.
Supporting information to this article can be found online at https://doi.org/10.1016/j.apsb.2025.07.035 .